INDEX

Explanations

looking into the mirror

New Auto-Interp

Top Features by Cosine Similarity

Configuration

Prompts (Dashboard)

10,000 prompts, 128 tokens each

Dataset (Dashboard)

lmsys/lmsys-chat-1m

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Beste

-0.12

ear

-0.11

 GetHashCode

-0.10

StackNavigator

-0.09

asal

-0.09

 Ferry

-0.09

 Dann

-0.09

å¾®ç¬ĳ

-0.09

ContentLoaded

-0.09

 ArgumentOutOfRangeException

-0.08

POSITIVE LOGITS

 mirror

0.16

 mirrors

0.14

 freak

0.13

 look

0.13

éı¡

0.13

 Look

0.13

 Mirror

0.13

Look

0.12

look

0.12

 self

0.11

Activations Density 0.027%